Li Ruize

Failure · The Sinking Prototype

The first prototype capsized within 30 seconds — poor waterproofing, motor short-circuited. Everyone went silent. But we didn't give up. We fished it out, took it apart, analyzed every leak point, and redesigned the seal. The second and third attempts still failed, but the fifth version finally floated. Failure isn't the end — it's the path to success.

Curiosity · Where It Began

Curiosity is the engine of all innovation — it drives us into the unknown.">It all started with a simple question: "Can robots help clean the ocean?" This question obsessed me. I devoured information on floating debris identification, studied ultrasonic sensor principles, and even taught myself basic image recognition. Curiosity is the engine of all innovation — it drives us into the unknown.

Creativity · Pioneering Spirit

No existing solution to copy, so we invented our own. Recycled plastic bottles for buoyancy, 3D-printed trash collection ports, combining infrared sensors with cameras to identify debris. No one on the team had built a water-surface robot before — every step was pioneering into the unknown, embodying MIT's "0 to 1" innovation ethos.

Risk-Taking · Dare to Try

Our mentor suggested a simpler land-based sorting robot — safer and easier. But we chose the harder path: a water-surface robot. Waterproofing, buoyancy, wave interference — each was a huge challenge. We actively chose the difficult road, because real innovation is never born in comfort zones. Embracing risk and uncertainty is essential for innovators.

Teamwork · 1+1+1+1>4

Four people, four personalities, four skills. I handled programming and sensors; Haifeng did mechanical design; Xiaoyu was our 3D modeling expert; Xiaojie ran testing and data logging. We had fierce arguments — like which waterproofing approach to use — but ultimately learned to listen, compromise, and trust. No one could do this alone. The team made each of us stronger.

Alignment with MIT's Mission

built ideas into reality with our own hands. From code to hardware, from theory to practice, every component united mind and hand. MIT seeks exactly these creators who can both think and build — using technology to solve real-world problems.">MIT's motto "Mind and Hand" — making knowledge useful. This project perfectly embodies that spirit: we didn't just theorize, we built ideas into reality with our own hands. From code to hardware, from theory to practice, every component united mind and hand. MIT seeks exactly these creators who can both think and build — using technology to solve real-world problems.

Social Responsibility

8 million tons of plastic enter the ocean each year. We built this project not just for fun, but because we genuinely want to do something for the ocean. One robot can only clean a tiny area, but it sends a message: technology should serve humanity and the planet. A true engineer isn't just someone who writes code — it's someone who cares about the world.

Whole Person · Resilience & Empathy

This project taught me more than technology. When teammates were frustrated, I learned to encourage them. When designs were rejected, I learned to accept criticism. When tests failed, I learned to manage emotions. A whole person matters more than perfect code — MIT doesn't just train engineers, it cultivates complete individuals with warmth, resilience, and empathy.

The project overlapped with final exams. I had to clearly prioritize: finish homework and study on school days, then fully commit to the project on weekends. Breaking big tasks into modules, setting milestones, distinguishing must-haves from nice-to-haves — learning to subtract is harder and more important than learning to add. Time is limited, but direction must be clear.

Tech Deep-Dive · Tesla FSD

Core: Tesla FSD uses a pure vision approach — 8 cameras + neural networks, no LiDAR. Transformer architecture stitches multi-camera feeds into a 360° bird's-eye view (Vector Space), real-time detecting lanes, vehicles, pedestrians, and traffic signals.

End-to-End: FSD v12+ uses end-to-end neural networks — camera input directly outputs steering and pedal commands, no hand-written rule code in between. Trained on millions of Tesla real-world driving clips, AI learns to drive like a human.

My Take: FSD proves a disruptive idea — with enough data, AI surpasses hand-written rules. This connects to my robot project: BlueGuardian also uses camera + AI to identify floating trash, not fixed rules. I want to dive deeper into computer vision and reinforcement learning to make robots perceive the world more intelligently.

Tech Deep-Dive · Starlink

Core: Starlink deploys thousands of small satellites in Low Earth Orbit (LEO, ~550km) to form a constellation for global high-speed internet. Traditional comm satellites at 35,786km (geostationary orbit) have ~600ms latency; Starlink's LEO reduces this to 20-40ms, nearly matching fiber-optic speeds.

Key Tech: ① Laser inter-satellite links — satellites communicate via laser without ground stations; ② Phased array antenna — user terminals track satellites electronically, no mechanical rotation; ③ Falcon 9 reusable rockets — launch cost is 1/10 of traditional, making mass deployment of thousands of satellites economically viable.

My Take: Starlink embodies the ultimate engineering mindset — not waiting for one perfect tech, but combining multiple technologies (reusable rockets + phased arrays + laser comms) to create a breakthrough. This inspires me: BlueGuardian can adopt modular thinking — optimize camera, sensors, and buoyancy separately, then combine into a stronger system. I want to learn more about communications and aerospace engineering.

Tech Deep-Dive · Falcon Rocket Recovery

Core: After stage separation, the Falcon 9 first stage reignites its engines to decelerate, deploys grid fins for aerodynamic control, and lands vertically on a drone ship (Of Course I Still Love You) or land pad — the first orbital-class rocket recovery and reuse in human history.

Failure Journey: From 2014 to 2016, SpaceX made nearly 10 landing attempts, many ending in explosions — CRS-7 broke apart mid-air in 2015, Amos-6 exploded during fueling in 2016, pushing the company to the brink of bankruptcy. But Musk said: "If things are not failing, you are not innovating enough." Every explosion was treated as precious data. The team iterated relentlessly until December 21, 2015 — the first successful landing. The control room erupted in cheers; Musk was speechless.

Admirable Spirit: In 2008, with both SpaceX and Tesla on the verge of bankruptcy, Musk put his last dollars into both companies. He said: "I'd rather fail building rockets than do nothing." This All-in courage, his calm acceptance of failure, and his unwavering belief in humanity's multi-planetary future — that's what moves me most.

My Take: Rocket recovery taught me that true innovation is inevitably accompanied by countless failures — and failure is not the end, but the path to success. In my BlueGuardian project, the robot also went through water leaks, sensor failures, and code crashes. Each failure brought me one step closer to success. Musk's story makes me more determined: no matter how hard the road ahead, I will bravely "explode" and keep moving forward — just like him.